Spanish researchers develop award-winning ancient manuscript recognition system
Spanish researchers have developed an award-winning system for the automatic recognition of manuscripts. The scientists, from the Autonomous University of Barcelona's Computer Vision Centre, have designed the efficient Blurred Shape Model (BSM) to be able to work with ancient, damaged or difficult to read manuscripts, handwritten scores and architectural drawings. According to the scientists, their research represents an effective human machine interface able to automatically reproduce documents while they are being written or drawn. The researchers won the first prize in the third edition of the Iberian Conference on Pattern Recognition and Image Analysis (IbPRIA) for their model. The researchers based their work on the biological process of the human mind and its ability to see and interpret all types of images (recognition of shapes, structures and dimensions) to create description and classification models of handwritten symbols. They say their computerised system differs from others in that it can detect variations, elastic deformations and uneven distortions that can appear when manually reproducing any type of symbol (letters, signs, drawings, etc.). Another advantage is the possibility to work in real time, only a few seconds after the document has been introduced into the computer. The researchers decided to test the efficiency of the system by experimenting in two application areas; a database of musical notes and a database of architectural symbols. The first included a total of 2,128 examples of three types of musical notes drawn up by 24 different people, whilst the second included 2,762 examples of handwritten architectural symbols belonging to 14 different groups. Each group contained approximately 200 types of symbols drawn up by 13 different people. The BSM was capable of recognising musical notes with an exactness of over 98% and architectural symbols with an exactness of 90%.
Countries
Spain